[openib-general] Re: [PATCH] mthca: correct max_rd_atomic handling
Michael S. Tsirkin
mst at mellanox.co.il
Thu Dec 15 08:43:37 PST 2005
Quoting r. Roland Dreier <rdreier at cisco.com>:
> Subject: Re: [PATCH] mthca: correct max_rd_atomic handling
>
> What happens if SAE and SRE are turned off and the consumer posts an
> RDMA read? Does it fail and generate an error completion?
Hardware guys confirmed that it does, as per spec: clearing these bits is
the way to tell hardware that we have max_rd_atomic set to 0.
I thought its obvious from documentation: do you think this needs clarification?
> I don't think we want it to just stop the QP processing.
It doesnt do that.
--
MST
More information about the general
mailing list